IBI-001404 - Severepublie
Word Error Rate Bias
Biais de WER (Word Error Rate)
IBI Score
7.2
out of 10
Severity4_grave
Frequencyfrequent
Scope groups4
Detection difficulty3
🇫🇷 Définition (FR)
Le WER masque les disparités de performance des systèmes de reconnaissance vocale selon les accents.
🇬🇧 Definition (EN)
WER masks disparities in speech recognition performance across accents.
TriAxis Taxonomy
Structural Classification
O
Origin
O2
Computational & Statistical
L
Lifecycle
L6
Evaluation
S
Specific
S22
Linguistic bias
💡
Concrete Example
Les systèmes de reconnaissance vocale ont un WER 2× plus élevé sur les accents noirs américains.
📊 Detection Metrics
Demographic Parity DifferenceDisparate Impact RatioEqualized Odds Difference
⚙ Alert Thresholds
Disparate Impact Ratio0.8
Equalized Odds Difference0.1
Demographic Parity Difference0.1
📚 Sources & Scientific References
- •Racial disparities in automated speech recognitionAllison Koenecke, Andrew Nam, Emily Lake, Joe Nudell, Minnie Quartey, Zion Mengesha, Connor Toups, John R. Rickford, Dan Jurafsky, Sharad Goel (2020)
- •
🛡 Impact & Mitigation
🎯 Vulnerable groups impacted
Black / Afro-descendantMinority language speakers
🛠 Mitigation methodologies
- AUDIT_TIERS : Third-party Audit
🏢 Sectors concerned
Media & Social Platforms (Risk: /5)Healthcare & Medicine (Risk: /5)
⚖ Applicable regulations
🇪🇺 Europe
- •AI_ACT_10 : AI Act (UE) 2024/1689
🌍 International
- •ISO_42001 : ISO/IEC 42001:2023
- •UNESCO_2021 : UNESCO - Recommandation sur l'éthique de l'IA
- •OECD_AI_PRINCIPLES : OCDE - Principes IA
- •COE_AI_CONVENTION_INT : Conseil de l'Europe - Convention IA
📋 Metadata
Contributor
Mohamadou SIDIBE
Institution
AI BIAIS INITIATIVE
Version
1
Created on
October 14, 2026
Updated on
November 4, 2026
UUID
8f299d10-9885-46a5-b997-9d75f4d82364
Explore other biases
250 documented biases, classified by the TriAxis taxonomy.